L
Linzi Xing
Publications - 15
Citations - 147
Linzi Xing is an academic researcher. The author has contributed to research in topics: Topic model & Computer science. The author has an hindex of 5, co-authored 13 publications receiving 86 citations.
Papers
More filters
Proceedings Article
Multilingual Twitter Corpus and Baselines for Evaluating Demographic Bias in Hate Speech Recognition
TL;DR: This work assemble and publish a multilingual Twitter corpus for the task of hate speech detection with inferred four author demographic factors: age, country, gender and race/ethnicity, and measures the performance of four popular document classifiers and evaluates the fairness and bias of the baseline classifiers on the author-level demographic attributes.
Proceedings ArticleDOI
Exploring Timelines of Confirmed Suicide Incidents Through Social Media
TL;DR: A novel dataset of Chinese social media accounts of 130 people who committed suicide between 2011 and 2016 is introduced, and a longitudinal text analysis of their post histories is conducted, showing observable changes in content leading up to the time of death.
Posted Content
Multilingual Twitter Corpus and Baselines for Evaluating Demographic Bias in Hate Speech Recognition
TL;DR: In this article, a multilingual Twitter corpus for the task of hate speech detection with inferred four author demographic factors: age, country, gender and race/ethnicity was assembled and published.
Proceedings ArticleDOI
Incorporating Metadata into Content-Based User Embeddings.
Linzi Xing,Michael J. Paul +1 more
TL;DR: This work proposes a data augmentation method that allows novel feature types to be used within off-the-shelf embedding models, and shows that this approach can lead to substantial performance gains with the simple addition of network and geographic features.
Proceedings Article
Diagnosing and Improving Topic Models by Analyzing Posterior Variability.
Linzi Xing,Michael J. Paul +1 more
TL;DR: This work proposes a metric called topic stability that measures the variability of the topic parameters under the posterior and shows that this metric is correlated with human judgments of topic quality as well as with the consistency of topics appearing across multiple models.